Mateusz Ochal

PhD student @EDINrobotics @InfAtEd @HeriotWattUni interested in #AI, #Robotics and #ComputerVision

[arXiv Daily @Arxiv_Daily]

How Sensitive are Meta-Learners to Dataset Imbalance?
https://t.co/WQxX7H34IB
by @OchalMateusz et al. including @mpatacch
#OpenSource #ComputerScience

[arXiv Daily @Arxiv_Daily]

Few-Shot Learning with Class Imbalance
https://t.co/Qnq1yHS2BG
by @OchalMateusz et al. including @mpatacch
#SupervisedLearning #OpenSource

[Edinburgh Robotics @EDINrobotics]

It was the night before Christmas in the Robotics lab. Find out if the robots saved the day when Santa couldn't cope!

[Mateusz Ochal @OchalMateusz]

Found it an interesting read on the quality of deep learning research papers https://t.co/Y6CUtyluFv . The reddit comments give some good paper examples

[/MachineLearning @slashML]

Examples of good science in deep learning experiment-focused papers https://t.co/8LJ4oLanwV

[Mateusz Ochal @OchalMateusz]

Strategyproof Reinforcement Learning for Online Resource Allocation https://t.co/u6X69sRnpy via @

[Antreas Antoniou @AntreasAntonio]

A youtube presentation of our paper "Defining Continual Few Shot Learning" hosted by the @ContinualAI group. https://t.co/8nOJEEf3hj

Paper information: https://t.co/RLkUVGrSBb

[Mateusz Ochal @OchalMateusz]

Check out our latest work exploring Few-Shot learning in a Continual learning context.

[Antreas Antoniou @AntreasAntonio]

Introducing some of our latest work on defining benchmarks for continual few-shot learning. Motivating continual few-shot learning and building a thorough suite of benchmarks for both meta and non-meta learning models.

Paper: https://t.co/3BIR3d3SoQ
Includes links to the repos

[BBC Scotland News @BBCScotlandNews]

"It's the first time I've seen the sun since I've been here".

George Clooney has a word about the Scottish weather as he and wife Amal attend a charity gala in Edinburgh.

[Najla Alariefy @NajlaAlariefy]

@AndrewYNg https://t.co/8NtHUOMO1I propose a solution for ML model management — much needed for empirical academic/industrial research. @Cometml

[Vladimir Iglovikov @viglovikov]

@AndrewYNg I am using https://t.co/ZZg2n3YNMt for versioning data and models in my DL pipelines.

[arXiv Daily @Arxiv_Daily]

Stay current on the latest #Learning #ComputerScience #ArtificialIntelligence #MachineLearning research published by @marcgbellemare this week https://t.co/JaJrgoZvHn

[arXiv Daily @Arxiv_Daily]

#PatternRecognition #ComputerVision https://t.co/iU4w7aZBdG

[arXiv Daily @Arxiv_Daily]

This one is a must read - the latest #MachineLearning #Learning #ArtificialIntelligence #ComputerScience research by @pabbeel @mooopan https://t.co/s3Vif0nrOx

[Andrew Ng @AndrewYNg]

Chatted about my journey through AI, including stories from early days of Google Brain & Coursera, plus Landing AI and https://t.co/Ryb1M2QyNn. Hope you'll find useful ideas and inspiration from this. Thanks @kevin_scott for doing this! https://t.co/dleWKkGYEA

[Google AI @GoogleAI]

Today, we are launching the What-If Tool, a new feature of the open-source TensorBoard web application, which let users analyze and better understand an ML model without writing code. We look forward to people using, and contributing to, the What-If Tool. https://t.co/s5ivnQH87q

[Ian Goodfellow @goodfellow_ian]

Congratulations to @julius_adebayo , @_beenkim, @muelly and @mrtz ! "Sanity Checks for Saliency Maps" (on arxiv soon) will be a spotlight at NIPS 2018

[Google AI @GoogleAI]

Announcing the launch of Dataset Search, a new way for researchers to find the datasets they need, wherever they’re hosted, whether it’s a publisher's site, a digital library, or an author's personal web page. Learn more at https://t.co/tpdOkjAVgW

[arXiv Daily @Arxiv_Daily]

#PatternRecognition #ComputerVision https://t.co/WSf0efip6j

[Ian Goodfellow @goodfellow_ian]

Train ImageNet in 18 minutes for just $40. https://t.co/CdLtIS6Y44 By my former colleague @yaroslavvb

[Google DeepMind @GoogleDeepMind]

Variational Bayesian Reinforcement Learning with Regret Bounds: https://t.co/JRfsJnm5nq